Religion in Latin America, 2014
CitationBell, J., & Sahgal, N. (2021, September 6). Religion in Latin America, 2014.
SummaryBetween October 2013 and February 2014, Pew Research Center, with generous funding from The Pew Charitable Trusts and the John Templeton Foundation, conducted a public opinion survey involving more than 30,000 face-to-face interviews in 19 countries and territories across Latin America and the Caribbean. This survey covers religious affiliations, beliefs, practices, social and political views in 18 countries and the U.S. territory (Puerto Rico). The survey was carried out as part of the Pew-Templeton Global Religious Futures project, which analyzes religious change and its impact on societies around the world.
The ARDA has added three additional variables to the original data set to enhance the users' experience on our site.
Data FileCases: 30326
Weight Variable: WEIGHT
Data were weighted to account for differences in probability of selection. Additionally, in some cases survey results were weighted to match the demographic characteristics of the population in each country. The weight variable is called "WEIGHT" and should be used in all analyses geared toward reporting marginal results. This variable does not combine all 18 countries into a single regional sample that accounts for population differences. It should only be used for analyses of national populations and subgroups of national populations.
Base weight and post-stratification weight combined in WEIGHT.
Range: 0.026 to 4.147
Data CollectionOctober 2013 - February 2014
Original Survey (Instrument)Religion in Latin America
Funded ByPew Charitable Trust
John Templeton Foundation
Collection ProceduresThe survey was conducted under the direction of Princeton Survey Research Associates International in Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua and Panama, and Ipsos Public Affairs in Argentina, Bolivia, Brazil, Chile, Colombia, the Dominican Republic, Ecuador, Paraguay, Peru, Puerto Rico, Uruguay and Venezuela. In all countries, surveys were administered through face-to-face interviews conducted at the respondent's place of residence.
The questionnaire administered by survey interviewers was designed by Pew Research Center staff in consultation with subject matter experts and advisers to the project. The questionnaire was translated into Spanish, Portuguese and Guarani, independently verified by professional linguists conversant in regional dialects and pretested prior to fieldwork.
Sampling ProceduresThe survey is based on samples of non-institutionalized adults ages 18 and older. In some instances, samples exclude minor segments of the population due to accessibility problems or security concerns. In Mexico, an oversample was conducted in the southern states of Chiapas, Tabasco, Campeche and Quintana Roo - states known to have higher concentrations of Protestant Christians.
All samples are based on multi-stage cluster designs, which typically entailed proportional stratification by region, locality size and urbanity, selection of primary sampling units (PSUs) proportional to population size, and random selection of secondary and tertiary sampling units within PSUs. Interview teams were assigned to designated random routes at the block or street level and followed predetermined skip patterns when contacting households. Within households, interviewers randomly selected a respondent by using a Kish grid (a random selection from a detailed list of all household members) or by selecting the adult with the next or most recent birthday.
Following fieldwork, survey performance for each country was assessed by comparing the results for key demographic variables with reliable, national-level population statistics. For each country, the data were weighted to account for different probabilities of selection among respondents. Where appropriate, data also were weighted through an iterative procedure to more closely align the samples with official population figures for gender, age and education. The reported sampling errors and the statistical tests of significance used in the analysis take into account the effects of weighting and specific sample designs.
It should be noted that practical difficulties in conducting multinational surveys can introduce potential error or bias into the findings. In some countries, fieldwork checks revealed that initial samples were disproportionally female. Further investigation indicated that female interviewers, in particular, had departed from established protocols in order to limit the number of evening hours spent in unfamiliar or threatening neighborhoods. This resulted in fewer interviews with men, who often were not at home during the daytime.
Following this discovery, interviewers in Bolivia, Colombia, the Dominican Republic, Ecuador, Puerto Rico and Venezuela were trained to follow a modified fieldwork protocol that prioritized the selection of men but maintained random selection of respondents. And in the Dominican Republic, Ecuador, Puerto Rico and Venezuela, additional interviews were conducted to reduce the gender skew of the samples. New primary and secondary sampling units were randomly selected for the additional interviews without affecting the overall structure of the sampling design. These corrective fieldwork measures substantially reduced, but did not entirely eliminate, the gender imbalance in the final samples for Bolivia, Colombia, the Dominican Republic, Ecuador, Puerto Rico and Venezuela. Statistical weights were applied to further narrow the gender gap in these countries. Post-field analysis of the survey data revealed that the disproportionate share of female respondents does not influence the substantive survey findings.
To protect respondent confidentiality, indicators geographic region (within countries) have been withheld from the dataset. Scholars who wish to use the suppressed variables in their research should contact
In order to adapt the survey to local sensitivities and conditions, not all survey questions were asked, or identically phrased, in all countries. Variable labels and value labels in the dataset may not reflect variations in question wording and response categories for each country. In addition to sampling error and other practical difficulties, one should bear in mind that question wording also can have an impact on the findings of opinion polls.
Principal InvestigatorsJames Bell and Neha Sahgal
Related PublicationsPew Research Center, Nov. 13, 2014, "Religion in Latin America: Widespread Change in a Historically